Themes, Saving and Interactive Plots

 

These exercise cover the scales, statistics and themes of ggplot2 for Plotting in R.

Exercise 1 - Themes

• Read in the cleaned patients dataset as we saw in ggplot2 course earlier (“patients_clean_ggplot2.txt”)

patients_clean <- read.delim("data/patients_clean_ggplot2.txt",sep="\t")

• Set the global theme to use theme_bw(). Using the patient data set generate a scatter plot of BMI versus Weight. Add a color scale to the scatter plot based on the Pet variable. Use an additional geom to add an extra layer of a fit line to our scatterplot (use lm method). Lets also add a nice color palette of your choice from Colorbrewer, Paleteer or Viridis.

library(ggplot2)

theme_set(theme_bw())

plot <- ggplot(data=patients_clean,
               mapping=aes(x=BMI,y=Weight,colour=factor(Pet)))+
  geom_point()+
  stat_smooth(method="lm",se=F) + scale_color_viridis_d(option="plasma")
plot

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Remove the legend title from the plot, change the background colors of legend panels to gray and place legend at bottom of plot.

plot <- plot + theme(legend.title=element_blank(),
                   legend.background=element_rect(fill="gray"),
                   legend.key=element_rect(fill="gray"),
                   legend.position="bottom")

plot

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Remove the legend title from the plot, change the background colors of legend panels to white and place legend at bottom of plot.

plot <- plot+facet_wrap(~Pet)

plot

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Add a title to the plot, remove minor grid lines and save to 7 by 7 inch png file.

plot <- plot+ggtitle("BMI vs Weight - By Pet")+theme(panel.grid.minor=element_blank())

plot

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

ggsave(plot,file="BMIvsWeight.png",units = "in",height = 7,width = 7)

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Save our current theme then try out some alternative themes. Try one theme included with ggplot2 and one from ggthemes.

myTheme <- plot$theme

plot + theme_linedraw()

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

library(ggthemes)
plot + theme_fivethirtyeight()

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Take at least one element from each theme you just tried out and add it to our existing theme.

myTheme <- myTheme + theme(strip.background = element_rect(fill = "black"), # from theme_linedraw
            strip.text = element_text(colour = "white", size = rel(0.8)), # from theme_linedraw
             panel.grid = element_line(colour = NULL), # from theme_fivethirtyeight
            panel.grid.major = element_line(colour = "#D2D2D2"), # from theme_fivethirtyeight
            panel.grid.minor = element_blank(),# from theme_fivethirtyeight
            rect = element_rect(fill = "#F0F0F0", # from theme_fivethirtyeight
            linetype = 0, colour = NA))# from theme_fivethirtyeight

• Use the + to update our plot to use the new theme

plot + myTheme

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Save the final plot as a pdf.

ggsave(plot, file="BMIvsWeight.pdf")

## Saving 7 x 5 in image
## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

• Reset the theme to the default: theme_gray()

theme_set(theme_gray())

plot

## `geom_smooth()` using formula = 'y ~ x'

## Warning: Removed 5 rows containing missing values or values outside the scale range
## (`geom_point()`).

Exercise 2 - External Packages

• Load in the data from “data/DESeq2_PCA.RData”. This contains a ggplot object that is produced by the package DESeq2. It is a PCA plot produced from differential gene expression analysis. Review the object. Check the plot, data, aesthetics etc.

load("data/DESeq2_PCA.RData")

## Warning: namespace 'DESeq2' is not available and has been replaced
## by .GlobalEnv when processing object 'deseq_pca_example'

deseq_pca_example

deseq_pca_example$data

##                PC1         PC2 group condition     name
## sample1  -13.24423  22.0512418     A      Ctrl  sample1
## sample2  -14.40362 -13.8514994     A      Ctrl  sample2
## sample3  -14.79663  -2.9299797     A      Ctrl  sample3
## sample4  -15.12221  -9.4008058     A       Mut  sample4
## sample5  -14.78719  -0.2914010     A       Mut  sample5
## sample6  -13.77777   5.6896212     A       Mut  sample6
## sample7   13.68499  -2.4219808     B      Ctrl  sample7
## sample8   14.85294   3.9328670     B      Ctrl  sample8
## sample9   12.21286  -3.0234083     B      Ctrl  sample9
## sample10  16.25675  -5.6924350     B       Mut sample10
## sample11  14.85222   0.9620356     B       Mut sample11
## sample12  14.27188   4.9757444     B       Mut sample12

deseq_pca_example$mapping

## Aesthetic mapping: 
## * `colour` -> `group`
## * `x`      -> `PC1`
## * `y`      -> `PC2`

• Lets change the color scheme and add a title.

deseq_pca_example + scale_fill_viridis_d() + ggtitle("PCA plot of DGE in A and B groups")

• When we look at the data column we can see there is an additional column contained in the metadata - condition. Lets add this to the plot.

deseq_replot <- deseq_pca_example
deseq_replot$layers <- NULL

deseq_replot <- deseq_replot + geom_point(aes(x=PC1, y=PC2, color=group, shape=condition)) + scale_fill_viridis_d() + ggtitle("PCA plot of DGE in A and B groups")

Exercise 3 - Interactive Plots

• Turn our PCA plot into an interactive plot.

library(plotly)

ggplotly(deseq_replot)

• Add the names of each sample from the orginal data object to the interactive labeling.

ggplotly(deseq_replot + geom_point(aes(label = name)))

## Warning in geom_point(aes(label = name)): Ignoring unknown aesthetics: label